redonkable
/
alistair23-linux
1
0
Fork 0
Code Releases Activity

ocfs2: multi node truncate fix 

Fix ocfs2_truncate_file() so that it forces a truncate_inode_pages() on all
interested nodes in all cases of a truncate(), not just allocation change.

Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
hifive-unleashed-5.1
Mark Fasheh 2006-03-16 15:06:37 -08:00
parent 6246b6128b
commit ab0920ce7e
1 changed files with 11 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
fs/ocfs2/file.c
View File

@ -260,6 +260,17 @@ static int ocfs2_truncate_file(struct inode *inode,
if (new_i_size == le64_to_cpu(fe->i_size))
goto bail;
/* This forces other nodes to sync and drop their pages. Do
* this even if we have a truncate without allocation change -
* ocfs2 cluster sizes can be much greater than page size, so
* we have to truncate them anyway. */
status = ocfs2_data_lock(inode, 1);
if (status < 0) {
mlog_errno(status);
goto bail;
}
ocfs2_data_unlock(inode, 1);
if (le32_to_cpu(fe->i_clusters) ==
ocfs2_clusters_for_bytes(osb->sb, new_i_size)) {
mlog(0, "fe->i_clusters = %u, so we do a simple truncate\n",
@ -272,14 +283,6 @@ static int ocfs2_truncate_file(struct inode *inode,
goto bail;
}
/* This forces other nodes to sync and drop their pages */
status = ocfs2_data_lock(inode, 1);
if (status < 0) {
mlog_errno(status);
goto bail;
}
ocfs2_data_unlock(inode, 1);
/* alright, we're going to need to do a full blown alloc size
* change. Orphan the inode so that recovery can complete the
* truncate if necessary. This does the task of marking